Lightning damage risk assessment calculation parameter extraction method and device

Abstract

The invention discloses a lightning damage risk assessment calculation parameter extraction method and device, and the method comprises the steps: obtaining the laser point cloud data of each instantiated power part, and classifying the laser point cloud data according to the equipment type of each power part; according to the classified insulator point cloud data, insulator chain distance parameters are calculated through a principal component analysis method; according to the classified tower-lightning conductor point cloud data, calculating height parameters of a tower lightning conductor through a cloth filtering algorithm; obtaining the relative position of a lead hanging point according to the classified lead tower point cloud data, and calculating the height parameter of each phase lead of the tower through a digital terrain model; and based on the insulator string distance parameter, the tower lightning conductor height parameter and the height parameter of each phase of lead of the tower, constructing a lightning damage risk assessment calculation parameter. According to the invention, the problem that the accuracy of lightning damage risk assessment calculation parameter acquisition is greatly reduced due to manual measurement and the influence on the accuracy of subsequent lightning damage risk assessment is large is solved.

Description

Technical Field

[0001] This invention relates to the field of lightning hazard risk assessment, and in particular to a method and apparatus for extracting calculation parameters for lightning hazard risk assessment. Background Technology

[0002] Lightning hazard risk assessment is a crucial component of safety protection for high-voltage transmission systems and is widely used in the planning, operation, and status monitoring of transmission lines. The key lies in the measurement of the calculation parameters for lightning hazard risk assessment, and the accuracy of these measurements directly affects the accuracy of the lightning hazard risk assessment.

[0003] However, in traditional technical approaches, the geometric parameters required for lightning damage risk assessment often rely on manual measurement, two-dimensional drawing annotation, or semi-automatic extraction based on empirical rules. These methods, due to their overly subjective judgment and limited understanding of the site structure, often result in significantly reduced accuracy in obtaining lightning damage risk assessment parameters, especially in mountainous environments where factors such as large terrain undulations, complex transmission structures, and significant differences in the height of conductor attachment points and lightning protection wire points exist. This negatively impacts the accuracy of subsequent lightning damage risk assessments.

[0004] Therefore, there is an urgent need for a method and apparatus for extracting calculation parameters for lightning damage risk assessment. Summary of the Invention

[0005] This invention provides a method and apparatus for extracting calculation parameters for lightning damage risk assessment, which solves the problem that the accuracy of obtaining calculation parameters for lightning damage risk assessment is greatly reduced due to manual measurement, thus having a significant impact on the accuracy of subsequent lightning damage risk assessment.

[0006] In a first aspect, this invention provides a method for extracting calculation parameters for lightning damage risk assessment. The method includes: acquiring laser point cloud data of various instantiated power components, and classifying the laser point cloud data according to the equipment type of each power component. The classified laser point cloud data includes insulator point cloud data, tower-lightning conductor point cloud data, and conductor tower point cloud data. Based on the classified insulator point cloud data, the method calculates the insulator string distance parameter using principal component analysis. Based on the classified tower-lightning conductor point cloud data, the method calculates the tower-lightning conductor height parameter using a material filtering algorithm. Based on the classified conductor tower point cloud data, the method obtains the relative position of the conductor suspension point and calculates the conductor height parameter of each phase of the tower using a digital terrain model. Based on the insulator string distance parameter, the tower-lightning conductor height parameter, and the conductor height parameter of each phase of the tower, the method constructs lightning damage risk assessment calculation parameters.

[0007] Optionally, laser point cloud data of each instantiated power component is acquired, and the laser point cloud data is classified according to the equipment type of each power component. Specifically, this includes: performing semantic segmentation on the laser point cloud data to extract a subset of point clouds with semantic labels; constructing an adjacency graph based on the point cloud subset using k-nearest neighbor search; dividing the laser point cloud data of the same equipment type into several connected graphs based on the adjacency graph through connected component analysis, with each connected graph corresponding to a single power device in the equipment type; and performing a removal operation on the connected graphs according to a preset method to obtain the classified laser point cloud data.

[0008] Optionally, based on the classified insulator point cloud data, the distance parameter of the insulator string is calculated using principal component analysis. Specifically, this includes: standardizing the insulator point cloud data to obtain a data matrix corresponding to each insulator; constructing a covariance matrix based on the data matrix; solving for the eigenvectors corresponding to the insulator point cloud data based on the covariance matrix; selecting a preset number of eigenvectors to construct the principal component data corresponding to the insulator point cloud data; projecting the insulator point cloud data onto the direction of the principal component data to obtain a projected point cloud, and calculating the distance between the two endpoints of the projected point cloud; using the distance between the two endpoints of the projected point cloud as the distance parameter of the insulator string.

[0009] Optionally, a covariance matrix is ​​constructed based on the data matrix, specifically by constructing the covariance matrix using the following formula:

[0010] ;

[0011] in, Let covariance matrix be the variance matrix. For data matrix, The number of points in the data matrix. It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

[0012] Optionally, based on the classified tower-lightning conductor point cloud data, the tower-lightning conductor height parameters are calculated using a material filtering algorithm. Specifically, this includes: calculating ground point cloud data in the tower-lightning conductor point cloud data using a material filtering algorithm; constructing a digital surface model based on the ground point cloud data; calculating the nearest neighbor point between the lightning conductor and the tower to obtain the lightning conductor attachment point; and calculating the lightning conductor height parameters based on the projection of the lightning conductor attachment point onto the digital surface model.

[0013] Optionally, the ground point cloud data in the tower-lightning wire point cloud data is calculated using a material filtering algorithm, specifically including: the material filtering algorithm is expressed by the following formula:

[0014] ;

[0015] in, Indicates that fabric particles are in Location at any given moment For the mass of the fabric particles, The external force acting on the fabric particles The internal forces acting on the fabric particles.

[0016] Optionally, based on the classified conductor tower point cloud data, the height parameters of each phase conductor of the tower are calculated using a digital terrain model. Specifically, this includes: calculating the nearest neighbor point between the conductor and the tower insulator to obtain the conductor suspension point; obtaining the relative position of the conductor suspension point based on the classified conductor tower point cloud data; and calculating the height parameters of each phase conductor of the tower based on the projection and relative position of the conductor suspension point on the digital surface model.

[0017] Optionally, the relative positions of conductor suspension points are obtained based on the classified conductor tower point cloud data. Specifically, this includes: obtaining the suspension point coordinates of each phase conductor on the insulator based on the conductor tower point cloud data; sorting the suspension point coordinates and defining the observer's perspective based on the sorting results; calculating the spatial relationship between the suspension point connection vector and the observation vector based on the observer's perspective; and obtaining the relative positions of the conductor suspension points in the vertical and horizontal directions through the spatial relationship.

[0018] In a second aspect, the present invention provides a device for extracting calculation parameters for lightning damage risk assessment. The device includes an acquisition module and a processing module, wherein...

[0019] The acquisition module is used to acquire laser point cloud data of each instantiated power component and classify the laser point cloud data according to the equipment type of each power component. The classified laser point cloud data includes insulator point cloud data, tower-lightning wire point cloud data, and conductor tower point cloud data.

[0020] The processing module is used to calculate the insulator string distance parameters using principal component analysis based on the classified insulator point cloud data; calculate the tower lightning conductor height parameters using a material filtering algorithm based on the classified tower-lightning conductor point cloud data; obtain the relative positions of conductor suspension points based on the classified conductor tower point cloud data, and calculate the conductor height parameters of each phase of the tower using a digital terrain model; and construct lightning damage risk assessment calculation parameters based on the insulator string distance parameters, tower lightning conductor height parameters, and tower conductor height parameters of each phase of the tower.

[0021] Optionally, the acquisition module acquires laser point cloud data for each instantiated power component and classifies the laser point cloud data according to the device type of each power component, specifically including:

[0022] Semantic segmentation processing is performed on the laser point cloud data to extract a subset of point clouds with semantic labels;

[0023] Based on the aforementioned point cloud subset, an adjacency graph is constructed using k-nearest neighbor search;

[0024] Based on the adjacency graph, the laser point cloud data of the same equipment type is divided into several connected graphs through connected component analysis, and one connected graph corresponds to a single power equipment in the equipment type;

[0025] The connected graph is removed according to a preset method to obtain the classified laser point cloud data.

[0026] Optionally, the processing module calculates the insulator string distance parameters based on the classified insulator point cloud data using principal component analysis, specifically including:

[0027] The insulator point cloud data is standardized to obtain a data matrix corresponding to each insulator;

[0028] Construct a covariance matrix based on the data matrix;

[0029] The feature vector corresponding to the insulator point cloud data is obtained by solving the covariance matrix.

[0030] Select a preset number of the feature vectors to construct the principal component data corresponding to the insulator point cloud data;

[0031] The insulator point cloud data is projected onto the direction of the principal component data to obtain the projected point cloud, and the distance between the two endpoints of the projected point cloud is calculated.

[0032] The distance between the two endpoints of the projected point cloud is used as the distance parameter of the insulator string.

[0033] Optionally, constructing the covariance matrix based on the data matrix specifically includes:

[0034] The covariance matrix is ​​constructed using the following formula:

[0035] ;

[0036] in, Let be the covariance matrix. For the data matrix, The number of points in the data matrix. It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

[0037] Optionally, the processing module calculates the tower-lightning conductor height parameters based on the classified tower-lightning conductor point cloud data using a material filtering algorithm, specifically including:

[0038] The ground point cloud data in the tower-lightning wire point cloud data is calculated using a fabric filtering algorithm;

[0039] A digital land surface model is constructed based on the ground point cloud data;

[0040] Calculate the nearest neighbor point between the lightning protection wire and the tower to obtain the lightning protection wire hanging point;

[0041] The height parameter of the lightning protection wire is calculated based on the projection of the lightning protection wire attachment point onto the digital surface model.

[0042] Optionally, the fabric filtering algorithm is expressed by the following formula:

[0043] ;

[0044] in, Indicates that fabric particles are in Location at any given moment The mass of the fabric particles. The external force acting on the fabric particles. The internal force acting on the fabric particles is denoted as .

[0045] Optionally, the processing module calculates the conductor height parameters for each phase of the tower using a digital terrain model based on the classified point cloud data of the conductor tower, specifically including:

[0046] Calculate the nearest neighbor points between the conductor and the tower insulator to obtain the conductor suspension points;

[0047] The relative positions of the conductor suspension points are obtained based on the classified point cloud data of the conductor towers.

[0048] The height parameters of each phase conductor of the tower are calculated based on the projection of the conductor suspension point on the digital surface model and the relative position.

[0049] Optionally, obtaining the relative position of the conductor suspension point based on the classified conductor tower point cloud data specifically includes:

[0050] Based on the point cloud data of the conductor tower, obtain the coordinates of the hanging points of each phase conductor on the insulator;

[0051] The coordinates of the hanging points are sorted, and the observer's perspective is defined based on the sorting results;

[0052] Calculate the spatial relationship between the hanging point connection vector and the observation vector based on the observer's perspective;

[0053] The relative positions of the wire hanging points in the vertical and horizontal directions are obtained through the spatial relationship. In a third aspect, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory stores instructions, the user interface and network interface are used for communication with other devices, and the processor executes the instructions stored in the memory to cause the electronic device to perform the method described in any of the foregoing descriptions.

[0054] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, the computer program being executed by a processor according to any of the methods described above.

[0055] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0056] 1. Acquire laser point cloud data for each instantiated power component and classify the laser point cloud data according to the equipment type of each power component; calculate the insulator string distance parameter using principal component analysis based on the classified insulator point cloud data; calculate the tower lightning conductor height parameter using a material filtering algorithm based on the classified tower-lightning conductor point cloud data; obtain the relative position of the conductor suspension point based on the classified conductor tower point cloud data, and calculate the conductor height parameter of each phase of the tower using a digital terrain model; construct lightning damage risk assessment calculation parameters based on the insulator string distance parameter, tower lightning conductor height parameter, and tower conductor height parameter of each phase of the tower. This process, by instantiating power component point clouds and classifying the point cloud data, allows for the calculation of different lightning damage risk assessment parameters according to different types of point cloud data using appropriate corresponding methods. This significantly improves the accuracy of lightning damage risk assessment parameter calculation and provides reliable input parameter support for subsequent lightning damage risk assessments.

[0057] 2. Perform semantic segmentation on the laser point cloud data to extract point cloud subsets with semantic labels; based on the point cloud subsets, construct an adjacency graph using k-nearest neighbor search; based on the adjacency graph, divide the laser point cloud data of the same equipment type into several connected graphs through connected component analysis, with each connected graph corresponding to a single power device in the equipment type; remove connected graphs according to a preset method to obtain classified laser point cloud data, thereby achieving accurate classification of the instance-level structure of power components, improving the geometric integrity and spatial independence of the point cloud classification results, and providing a clear and identifiable input basis for subsequent extraction of structural parameters and spatial location association of each power device.

[0058] 3. Based on the point cloud data of the conductor tower, obtain the coordinates of the suspension points of each phase conductor on the insulator; sort the suspension point coordinates and define the observer's perspective based on the sorting results; calculate the spatial relationship between the suspension point connection vector and the observation vector based on the observer's perspective; obtain the relative positions of the conductor suspension points in the vertical and horizontal directions through the spatial relationship, thereby realizing the high, medium and low layer identification of conductor suspension points and the determination of left and right phases, clarifying the arrangement relationship of each conductor instance in three-dimensional space, and providing a structural basis for the accurate attribution of conductor height parameters, phase distance calculation, and phase correlation analysis in lightning damage risk assessment. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating a method for extracting calculation parameters for lightning damage risk assessment provided in an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of a method for determining the relative position of a conductor hanging point according to an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of a lightning damage risk assessment calculation parameter extraction device provided in an embodiment of the present invention;

[0062] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0063] Explanation of reference numerals in the attached drawings: 31, acquisition module; 32, processing module; 301, processor; 402, communication bus; 403, user interface; 404, network interface; 405, memory. Detailed Implementation

[0064] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0065] The terminology used in the following embodiments of the present invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the invention, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in the invention refers to and includes any or all possible combinations of one or more of the listed items.

[0066] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of the present invention, unless otherwise stated, "a plurality of" means two or more.

[0067] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0068] Please refer to Figure 1 The diagram illustrates a flowchart of a method for extracting calculation parameters for lightning damage risk assessment provided by an embodiment of the present invention. The flowchart mainly includes the following steps: S101 to S105.

[0069] Step S101: Obtain the laser point cloud data of each instantiated power component, and classify the laser point cloud data according to the equipment type of each power component.

[0070] Specifically, given the complex terrain and significant elevation differences in mountainous areas, laser point cloud data can be acquired using airborne lidar (ALS) systems. These systems are typically mounted on fixed-wing aircraft or rotary-wing UAVs, periodically emitting laser pulses towards the ground using a laser emitter. The reflection time is measured, and combined with an inertial navigation system (INS) and a global positioning system (GPS) for three-dimensional positioning, thereby obtaining high-resolution spatial point cloud data. To improve penetration in forested mountainous areas, lidar equipment with longer wavelengths and higher pulse frequencies is preferred, and combined with flight path optimization and multiple echo recording techniques to obtain more complete point clouds of the ground and power facilities.

[0071] Secondly, local supplementation can be achieved by combining terrestrial laser scanning (TLS). TLS is typically deployed at substation entrances and exits, tower bases, or areas with open views. It uses a high-precision rotating laser scanner to perform fixed-point 3D modeling, making it suitable for high-density, high-precision supplementation scenarios, such as detailed structural supplementation of insulators, clamps, or conductor suspension parts.

[0072] After point cloud acquisition, the raw point cloud data needs to be preprocessed. Preprocessing methods include, but are not limited to, echo selection, noise removal, coordinate transformation, intensity normalization, and timestamp synchronization. After preprocessing, the point cloud is classified point by point using a deep convolutional neural network or a point cloud semantic segmentation model based on multi-scale feature extraction (such as PointNet++ or RandLA-Net). The points are labeled with semantic tags such as "pole / tower," "conductor," "lightning conductor," and "insulator" according to the equipment type of the power components, forming a semantic-level point cloud structure. The classified laser point cloud data includes, but is not limited to, insulator point cloud data, pole / tower-lightning conductor point cloud data, and conductor / tower point cloud data.

[0073] In one possible implementation, step S101 further includes: performing semantic segmentation processing on the laser point cloud data to extract a subset of point clouds with semantic labels; constructing an adjacency graph based on the point cloud subset using k-nearest neighbor search; dividing the laser point cloud data of the same device type into several connected graphs based on the adjacency graph through connected component analysis, with each connected graph corresponding to a single power device in the device type; and performing a removal operation on the connected graphs according to a preset method to obtain the classified laser point cloud data.

[0074] Specifically, although the laser point cloud data of transmission lines after semantic segmentation assigns a semantic label to each point, it does not identify insulators, towers, or conductors as individual instances, making it impossible to extract parameters for specific towers, conductors, or insulators. Taking a point cloud sample of a single span of transmission lines as an example, for point clouds in this sample that have already been labeled "tower" or "insulator," we still need to assign them instance labels to distinguish between two different towers and different insulators.

[0075] Taking poles as an example, if we extract the point cloud with the semantic label "pole", considering that there are obvious spatial separations between poles, while the point cloud of a single pole is connected, we can use this characteristic to instantiate point clouds with the same semantic label according to their connectivity.

[0076] Connectivity analysis can be used to separate the point cloud of the transmission line itself from the ground surface points, but this is computationally inefficient for tens of millions of points. However, after semantic segmentation, we extract the point cloud belonging to the transmission line itself, which consists of only tens of thousands of points, making connectivity analysis of the point cloud much more efficient. The steps for instantiating power facilities or components using connectivity analysis are as follows:

[0077] First, semantic segmentation is performed on the laser point cloud data to extract a subset of points with semantic labels. The purpose of semantic segmentation is to assign a semantic category label to each point in the point cloud, with typical categories including towers, conductors, lightning protection wires, and insulators. This processing can be implemented based on deep learning models, commonly using native point cloud networks such as PointNet, PointNet++, KPConv, and RandLA-Net to perform feature encoding and category prediction on the input 3D point data, thereby generating refined point-level semantic annotation results. Through semantic segmentation, valid points related to power facilities can be extracted from the original point cloud, while interfering points such as vegetation, terrain, and buildings can be removed.

[0078] Secondly, based on the aforementioned point cloud subset, an adjacency graph is constructed using k-nearest neighbor search. k-nearest neighbor search is a local connectivity mechanism used to establish connections between each point and its spatial neighbors. In this implementation, a fixed value of k (e.g., k=10) is selected, and for each point, its k nearest neighbors in Euclidean space are retrieved. Adjacency relationships between these neighbors are then constructed, forming the point cloud graph structure. This adjacency graph not only preserves the semantic attributes of the points but also describes the spatial connectivity between them, providing support for subsequent graph algorithm analysis.

[0079] After the adjacency graph is constructed, connectivity analysis is performed based on this graph structure. The goal of connectivity analysis is to further refine the point cloud into structural units according to its spatial connectivity, that is, to identify spatially connected sets of points while maintaining semantic consistency. Connectivity analysis divides all points in the graph structure that can be connected by edge paths into a connected graph, and treats each connected graph as an individual electrical component under the device type. For example, point cloud data with the semantic label "insulator" will be divided into several connected graphs, each of which represents a physical string of insulators. This process achieves instance-level structure recognition above the semantic level, which is a key step from category recognition to structure modeling.

[0080] Subsequently, the connected graphs are culled according to a preset method. Because point cloud data may contain flying points, noise points, or small, incompletely covered false positive regions, some connected graphs have too few points to form a valid power facility geometry. Therefore, filtering is required based on a point count threshold (e.g., n=10). Any connected graph with fewer than this threshold will be removed to ensure that the final retained point cloud instances are complete, representative, and of engineering significance.

[0081] Finally, after completing the above series of processes, the classified laser point cloud data is obtained. This data not only has semantic category annotations but also structural instance segmentation under each category, forming a structured input dataset that meets the needs of subsequent parameter extraction. This provides a high-quality foundation for extracting parameters such as insulator string length, lightning conductor height, and conductor suspension point phase. This implementation method significantly enhances the organization, accuracy, and structural expressiveness of the point cloud data.

[0082] Furthermore, after instantiation based on connected component analysis, in order to associate the parameters extracted from each power facility or component with the tower number in the ledger, it is also necessary to add a tower number attribute to each power facility or component instance. The LAS format point cloud stores the coordinate system of the point cloud, while the ledger records the latitude and longitude coordinates of the towers. Each instance can be associated with the tower number through spatial positioning. The 220kV point cloud data of Yunnan region involved in this paper uses the WGS 1984 UTM 47N projected coordinate system. Therefore, for the point cloud data of a transmission corridor within a span (containing two towers), the latitude and longitude coordinates of the towers recorded in the ledger are first... After projection, it is converted to WGS 1984 UTM 47N projected coordinates. ), then calculate the instance. center of mass Projection on the xy plane With the projected coordinates of the two towers ( The distance is used to assign the tower number to the relevant instance if the tower is closer. .

[0083] Step S102: Based on the classified insulator point cloud data, calculate the insulator string distance parameters using principal component analysis.

[0084] Specifically, after point cloud semantic segmentation and instantiation, each insulator can be segmented individually. When the point cloud density is low, some insulator point cloud data may be missing. If the insulator string length is extracted by calculating the distance between the farthest point pairs in the insulator point cloud data, the line connecting the farthest point pairs often deviates significantly from the true axis of the insulator. To obtain the true direction of the insulator string point cloud data, this invention uses principal component analysis to fit the axis of the insulator string and calculates its length along the axis.

[0085] In one possible implementation, step S102 further includes: standardizing the insulator point cloud data to obtain a data matrix corresponding to each insulator; constructing a covariance matrix based on the data matrix; solving for the eigenvectors corresponding to the insulator point cloud data based on the covariance matrix; selecting a preset number of eigenvectors to construct principal component data corresponding to the insulator point cloud data; projecting the insulator point cloud data onto the principal component data direction to obtain a projected point cloud, and calculating the distance between the two endpoints of the projected point cloud; and using the distance between the two endpoints of the projected point cloud as the insulator string distance parameter.

[0086] Specifically, Principal Component Analysis (PCA) is a statistical method that uses orthogonal transformations to convert a set of potentially correlated observations into a set of linearly uncorrelated variables; this new set of variables is called the principal components. The goal of PCA is to reduce the dimensionality of the data while preserving most of its variability. Calculating the length of an insulator string using PCA involves the following steps:

[0087] In one possible implementation, step S102 further includes: performing axis fitting and length extraction processing on the insulator point cloud data based on principal component analysis (PCA) to improve the stability and accuracy of insulator geometric parameter calculation when point cloud density is uneven or partially missing. Specifically, this includes the following steps:

[0088] First, the insulator point cloud data is standardized to eliminate the impact of scale differences across dimensions on subsequent covariance calculations. Standardization is achieved by subtracting the mean from each coordinate dimension and dividing by its standard deviation, resulting in a data matrix with a mean of zero and a variance of one. This standardized data matrix constitutes the data matrix for each insulator point cloud. The size is n×3, where n is the number of points.

[0089] Then, a covariance matrix C is constructed based on the data matrix to quantify the joint variation relationships of the point cloud across different dimensions. The calculation method is as follows:

[0090] ;

[0091] in, Let covariance matrix be the variance matrix. For data matrix, It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

[0092] Next, the corresponding eigenvectors and eigenvalues ​​are calculated based on the covariance matrix. (Covariance matrix) It is A symmetric matrix of size is subjected to eigenvalue decomposition using the following formula:

[0093]

[0094] in, For eigenvalues, These are the feature vectors. Different feature vectors can then be obtained. and corresponding eigenvalues The eigenvectors are sorted according to the magnitude of their eigenvalues. The larger the eigenvalue, the greater the variance of the data in that direction, meaning the more important the feature.

[0095] Subsequently, a preset number of feature vectors are selected in descending order of feature values, and the first k feature vectors are selected as the principal component data corresponding to the insulator point cloud data.

[0096] Projecting the insulator point cloud onto the direction of the principal component data yields a one-dimensional projected point cloud. The distance between the minimum and maximum projected values ​​in the projected point cloud is then calculated, representing the length of the insulator string along its principal axis. The relationship between the length of the insulator string along its principal axis and the distance between the two endpoints of the projected point cloud is essentially a process of converting a geometric projection in a high-dimensional space into an interval measurement in a one-dimensional coordinate system. Its mathematical and geometric logic is as follows: The insulator string point cloud is a discrete set of points composed of several three-dimensional coordinate points, linearly distributed along the principal axis of the insulator in three-dimensional space. Principal component analysis can extract the first principal component vector from the covariance matrix of this point set. This vector represents the direction of greatest variation in space, which is also the geometric principal axis direction of the insulator string in its structure. Construct a straight line passing through the center of the point cloud in this direction, and orthogonally project all three-dimensional coordinate points onto this direction vector. That is, the projected coordinate value of each point is the dot product of that point and the unit vector of the principal axis direction. The result is a one-dimensional projection value, which constitutes a one-dimensional projected point cloud. This one-dimensional projected point cloud can be regarded as the "shadow" of the original three-dimensional point cloud in the principal axis direction of the insulator, that is, the distribution coordinates of all original points in this direction. Since the principal axis direction is the extension direction of the dominant structure of the point cloud, the coverage of the projected point cloud in this direction can directly reflect the actual spatial length of the insulator string in this direction.

[0097] The method for obtaining the projected point cloud includes the following steps: Principal component analysis (PCA) is used to extract the first principal component vector from the 3D coordinate point cloud of each individual insulator. This first principal component vector represents the maximum variation trend of the individual insulator along its principal axis in 3D space, which is usually consistent with the longitudinal direction of the insulator string. To obtain this direction, the insulator point cloud data needs to be normalized by standard deviation, i.e., subtracting the mean from each dimension and then dividing by its standard deviation to eliminate the influence of scale differences. Next, a 3D covariance matrix is ​​constructed, and eigenvalue decomposition is performed on the covariance matrix to obtain three eigenvectors corresponding to the three eigenvalues. These eigenvectors are then sorted from largest to smallest, with the eigenvector corresponding to the largest eigenvalue being the unit vector along the principal axis of the insulator. Based on this principal axis, all points in the insulator point cloud are vector-projected, i.e., the one-dimensional projection value along the first principal component direction is calculated for each 3D point coordinate, thereby compressing the 3D point cloud into a one-dimensional projected point cloud. After obtaining the projected point cloud, the difference between the maximum and minimum values ​​among all projected values ​​is calculated. This difference represents the range length of the insulator point cloud along its principal axis, which corresponds to the geometric length of the insulator string along its principal axis. Essentially, this length calculation process reduces the dimensionality of the original point cloud data to the insulator's principal axis direction using principal component analysis and uses its range to characterize the spatial extension range along that direction. This method does not rely on point cloud density or structural integrity, exhibiting good data adaptability and geometric stability, and is particularly suitable for regularly shaped power components extending along a single axis, such as insulator strings. After standard deviation normalization, a covariance matrix is ​​constructed, and eigenvalue decomposition is performed. The eigenvector corresponding to the largest eigenvalue is selected as the principal axis direction. Subsequently, a vector projection operation is performed on the 3D point cloud along this direction, resulting in a one-dimensional projection value sequence. This one-dimensional projection value sequence is the projected point cloud, and the difference between its first and last endpoints represents the insulator string length. This method not only reflects the actual geometric length but also possesses robustness and scalability, making it suitable for batch processing and string length extraction tasks of multiple insulator individuals in large-scale point cloud data.

[0098] Finally, the distance between these two endpoints is used as the insulator string distance parameter to characterize the axial dimensional features of this insulator instance. This parameter provides crucial support for subsequent tower structure modeling, electrical clearance calculation, and lightning damage risk assessment.

[0099] Step S103: Based on the classified tower-lightning wire point cloud data, calculate the tower-lightning wire height parameters using a material filtering algorithm.

[0100] Specifically, the height of the lightning protection wire on the tower is the height of the suspension point of the lightning protection wire on the tower from the ground. The extraction of this parameter involves two key points: first, obtaining the ground model, and second, determining the suspension point.

[0101] In one possible implementation, step S103 further includes: calculating ground point cloud data in the tower-lightning wire point cloud data using a cloth filtering algorithm; constructing a digital surface model based on the ground point cloud data; calculating the nearest neighbor point between the lightning wire and the tower to obtain the lightning wire attachment point; and calculating the lightning wire height parameter based on the projection of the lightning wire attachment point onto the digital surface model.

[0102] Specifically, to calculate the height of the lightning protection wire hanging point from the ground, it is first necessary to effectively extract the ground point cloud. This paper uses the Cloth Simulation Filter (CSF) algorithm to extract the ground point cloud of a 220kV mountainous area in Yunnan. The Cloth Simulation Filter algorithm is an algorithm used to separate ground points and non-ground points (such as buildings, vegetation, etc.) from a point cloud. It is inspired by the real-world phenomenon of cloth covering the surface of objects and identifies the ground by simulating the process of cloth falling. Traditional ground point cloud extraction algorithms often consider changes in the elevation or slope of the point cloud. However, the authors of the Cloth Simulation Filter algorithm propose to invert the point cloud and then assume that a piece of cloth composed of multiple "particles" falls from above onto the ground points of the point cloud. Points in the point cloud that are less than a pre-set threshold from this piece of cloth are considered ground points. The Cloth Simulation Filter algorithm is expressed by the following formula:

[0103] ;

[0104] in, Indicates that fabric particles are in Location at any given moment For the mass of the fabric particles, The external force acting on the fabric particles The internal forces acting on the fabric particles.

[0105] The power lines involved in this invention are all located in mountainous areas with significant variations in ground slope. Therefore, the "cloth rigidity" is set to a smaller value to improve the cloth filtering algorithm's extraction effect on ground point clouds in power transmission corridors. Furthermore, due to the dense vegetation in mountainous areas, many ground points are covered by vegetation or buildings during airborne lidar scanning. Therefore, the distance threshold for ground point extraction is set to 1m, the cloth particle grid resolution is 0.5m, and the number of iterations is set to 800.

[0106] After extracting ground points, the ground point cloud often contains "holes" due to obstruction by vegetation and buildings. Therefore, it is necessary to generate a Digital Terrain Model (DTM) from the ground points to more accurately calculate the distance from the lightning rod to the ground. To balance computational efficiency and accuracy, Kriging interpolation is used to calculate the distance from the ground to the ground. The formula for a digital terrain model with a 1m grid precision is as follows:

[0107] ;

[0108] in, The terrain elevation value at location The estimated value, The number of known points. It is the first The values ​​of known points. Kriging interpolation is a statistically based best linear unbiased estimation method. Its basic idea is that spatially close things are more likely to have similar attribute values. Compared with other interpolation methods such as nearest neighbor, linear interpolation, or spline interpolation, Kriging interpolation has the advantage of more detailed analysis and utilization of the spatial correlation of data, considering not only the distance between points, but also the directionality and anisotropy of the spatial structure.

[0109] Next, the nearest neighbor point between the lightning protection wire and the tower is calculated to obtain the lightning protection wire attachment point, and the height parameter of the lightning protection wire is calculated based on the projection of the lightning protection wire attachment point on the digital surface model.

[0110] Step S104: Obtain the relative position of the conductor suspension point based on the classified conductor tower point cloud data, and calculate the conductor height parameters of each phase of the tower using a digital terrain model.

[0111] Specifically, the coordinates of the suspension points of each phase conductor on the insulator are obtained; the suspension point coordinates are sorted to determine the high, medium and low crossarms and their left and right positions; based on the projection position of the conductor suspension points in the digital terrain model, the difference between the suspension point elevation and the corresponding terrain elevation is calculated to obtain the height parameters of each phase conductor.

[0112] In one possible implementation, step S104 further includes: calculating the nearest neighbor points of the conductor and the tower insulator to obtain conductor suspension points; obtaining the relative positions of the conductor suspension points based on the classified conductor-to-tower point cloud data, specifically including: obtaining the suspension point coordinates of each phase conductor on the insulator based on the conductor-tower point cloud data; sorting the suspension point coordinates and defining the observer's perspective based on the sorting results; calculating the spatial relationship between the suspension point connection vector and the observation vector based on the observer's perspective; obtaining the relative positions of the conductor suspension points in the vertical and horizontal directions through the spatial relationship; and calculating the height parameters of each phase conductor of the tower based on the projection and relative position of the conductor suspension points on the digital surface model.

[0113] Specifically, based on the classified conductor point cloud and insulator point cloud data, a nearest neighbor search algorithm is used to calculate the minimum distance point pair between each conductor point cloud instance and the insulator point cloud instance on its corresponding tower. The conductor point corresponding to each pair of points with the minimum distance is taken as the attachment point of the conductor on the insulator. This attachment point is regarded as the physical contact point connecting the conductor and the tower structure, and is a key reference point for subsequent spatial positioning and elevation calculation.

[0114] Spatial relative position analysis was performed on all acquired conductor attachment points. The z-coordinates of all attachment points were extracted and sorted in ascending order. Based on the height stratification, the attachment points were divided into three layers: "high," "medium," and "low." Attachments with small differences in adjacent z-values ​​were grouped into crossarm pairs. Furthermore, by calculating the tower's orientation vector as the observer's perspective, the dot product relationship between the line vector connecting each attachment point pair and the perpendicular vector obtained by rotating the orientation vector 90 degrees clockwise was calculated to determine the left-right attribute of each attachment point in space. This process achieves a systematic classification of the conductor's spatial phase (high, medium, low) and position (left-right).

[0115] For example, taking a 220kV double-circuit transmission line with a drum-shaped tower as an example, there are a total of 6 transmission conductors in both circuits. Similar to the method for extracting the height of lightning protection wires, for each conductor instance, the closest point between the conductor instance and the tower insulator is calculated, and the distance from that point to the digital terrain model of the ground surface is extracted as the height of that conductor instance. So far, we have only extracted the height of each conductor instance, but we cannot know which line each conductor belongs to. We need a simple and easy-to-implement method to determine the relative position attributes of each conductor instance in space. Therefore, the problem to be solved now can be described as: for a point cloud sample of a transmission corridor with a span of one span, which includes tower point cloud instances... and The conductors of each phase have been located. Six hanging points on the insulator , The coordinates are How to identify the high, medium, low, left, and right attributes of the hanging points? For 6 hanging points... Sort the coordinates and determine their high, medium, low, left, and right positions. This is because there are two hanging points on the same crossarm. The coordinate values ​​are relatively close, assuming the sorting method is... Then, the hanging points 0 and 1, 2 and 3, 4 and 5 belong to the same crossarm, and thus a vector can be constructed. , , Assume that the tower is point to The vector is Then the direction pointed to by this vector is the direction of the line, which is also the direction from the observer's perspective. Assume one pair of conductors is suspended at... For vectors .Will The vector can be obtained by rotating it 90 degrees clockwise. This vector specifies the "right direction" along the path, and is calculated... and If the dot product is greater than 0, then the dot product... On the right; if less than 0, then On the right. Based on the above algorithm, the spatial relative position of the conductor suspension point can be identified; similarly, the relative position of the lightning protection wire can be determined. Please refer to... Figure 2 The diagram illustrates a method for determining the relative position of a wire hanging point according to an embodiment of the present invention.

[0116] Finally, based on the coordinates of each conductor attachment point in three-dimensional space, its projection position on the digital terrain model is extracted, and the corresponding elevation value of the projection point in the digital terrain model is found. The z-coordinate of the attachment point is subtracted from the elevation of the projection point in the digital terrain model to obtain the vertical height difference of the attachment point relative to the ground. Combined with the previously identified relative position attributes, this height difference is calibrated as the conductor height parameter corresponding to a specific phase and azimuth, thereby realizing the structured extraction of the conductor height of each phase of the tower.

[0117] Step S105: Based on the insulator string distance parameter, tower lightning protection wire height parameter, and tower phase conductor height parameter, construct lightning damage risk assessment calculation parameters.

[0118] Specifically, the insulator string length is used to determine the effective electrical insulation gap, and the height difference between the conductor and the lightning protection wire is used to analyze whether the protection range meets the requirements of the regulations. Combined with the phase arrangement and vertical spacing of the conductors, a parameter set reflecting the characteristics of the spatial lightning protection structure is constructed for subsequent quantitative assessment of risks such as lightning strike shielding failure and insufficient electrical clearance. Furthermore, the lightning damage risk assessment calculation parameters in the embodiments of this invention are not limited to the aforementioned insulator string distance parameters, tower lightning protection wire height parameters, and tower phase conductor height parameters. Other lightning damage risk assessment calculation parameters are obtained as follows:

[0119] Lightning conductor and phase conductor center distance parameters: For a complete 220kV drum-shaped tower point cloud, its height The direction changes the most, therefore this direction in PCA is either the direction of the first eigenvector or its opposite direction; its width... Change of direction is secondary, length is secondary. The direction is the direction of the second eigenvector of the PCA or its opposite direction; in Minimal change in direction, depth The direction is the direction of the third eigenvector of the PCA or its opposite direction. Therefore, the normal vector of the plane of symmetry used by the lightning conductor and each phase conductor can be calculated using the following formula:

[0120] ;

[0121] in, Let be the normal vector of the plane of symmetry. Let the principal component vectors of the point cloud be located in the width direction. Let be the principal component vector of the point cloud in the depth direction. Based on the point cloud... The midpoint of the directional projection is used to determine the position of the symmetry plane. Through the above steps, the symmetry plane of the tower point cloud can be obtained, and the center-distance parameters between the lightning protection wire and each phase conductor can be calculated using the point-to-plane distance formula.

[0122] Tower height parameter: Obtain the insulators connected to each conductor and map their spatial relative positions to the connected insulator instances. This will allow you to find the insulator suspended on the lowest crossarm of the tower. Find the nearest point of the tower corresponding to the two lower crossarm insulators and calculate the height above the ground using the digital ground model obtained in step S103. Take the maximum height as the tower height parameter.

[0123] Span parameter: Calculate the centroids of the two towers, and the projection distance of the line connecting the two centroids onto the xy plane is the span parameter.

[0124] Determining whether a tower is a tension tower or a straight tower: Associate all instances, including jumpers, with the tower number. Initially identify towers with drainage lines as tension towers, otherwise identify them as straight towers.

[0125] This invention acquires laser point cloud data of various instantiated power components using the aforementioned method, and classifies the laser point cloud data according to the equipment type of each power component. Based on the classified insulator point cloud data, principal component analysis is used to calculate the insulator string distance parameter. Based on the classified tower-lightning conductor point cloud data, a material filtering algorithm is used to calculate the tower-lightning conductor height parameter. Based on the classified conductor tower point cloud data, the relative position of the conductor suspension point is obtained, and the conductor height parameter of each phase of the tower is calculated using a digital terrain model. Based on the insulator string distance parameter, tower-lightning conductor height parameter, and tower-conductor height parameter of each phase of the tower, lightning damage risk assessment calculation parameters are constructed. By instantiating power component point clouds and classifying the point cloud data, different lightning damage risk assessment calculation parameters are calculated according to different types of point cloud data using appropriate corresponding methods. This greatly improves the accuracy of lightning damage risk assessment calculation parameter calculation and provides reliable input parameter support for subsequent lightning damage risk assessment.

[0126] Please refer to Figure 3 This illustration shows a schematic diagram of a lightning damage risk assessment calculation parameter extraction device provided in an embodiment of the present invention. The device includes an acquisition module 31 and a processing module 32, wherein...

[0127] The acquisition module 31 is used to acquire laser point cloud data of each instantiated power component and classify the laser point cloud data according to the equipment type of each power component. The classified laser point cloud data includes insulator point cloud data, tower-lightning wire point cloud data and conductor tower point cloud data.

[0128] Processing module 32 is used to calculate the insulator string distance parameters using principal component analysis based on the classified insulator point cloud data; calculate the tower lightning conductor height parameters using a material filtering algorithm based on the classified tower-lightning conductor point cloud data; obtain the relative positions of conductor suspension points based on the classified conductor tower point cloud data, and calculate the conductor height parameters of each phase of the tower using a digital terrain model; and construct lightning damage risk assessment calculation parameters based on the insulator string distance parameters, tower lightning conductor height parameters, and tower conductor height parameters of each phase of the tower.

[0129] In one possible implementation, the acquisition module 31 is used to acquire laser point cloud data of each instantiated power component and classify the laser point cloud data according to the equipment type of each power component. Specifically, this includes: performing semantic segmentation processing on the laser point cloud data to extract a subset of point clouds with semantic labels; constructing an adjacency graph based on the point cloud subset using k-nearest neighbor search; dividing the laser point cloud data of the same equipment type into several connected graphs based on the adjacency graph through connected component analysis, with each connected graph corresponding to a single power device in the equipment type; and performing a removal operation on the connected graphs according to a preset method to obtain the classified laser point cloud data.

[0130] In one possible implementation, the processing module 32 is used to calculate the insulator string distance parameter based on the classified insulator point cloud data using principal component analysis. Specifically, this includes: standardizing the insulator point cloud data to obtain a data matrix corresponding to each insulator; constructing a covariance matrix based on the data matrix; solving for the eigenvectors corresponding to the insulator point cloud data based on the covariance matrix; selecting a preset number of eigenvectors to construct the principal component data corresponding to the insulator point cloud data; projecting the insulator point cloud data onto the direction of the principal component data to obtain a projected point cloud, and calculating the distance between the two endpoints of the projected point cloud; and using the distance between the two endpoints of the projected point cloud as the insulator string distance parameter.

[0131] In one possible implementation, the processing module 32 is used to construct a covariance matrix based on the data matrix, specifically including: constructing the covariance matrix using the following formula:

[0132] ;

[0133] in, Let covariance matrix be the variance matrix. For data matrix, It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

[0134] In one possible implementation, the processing module 32 is used to calculate the height parameters of the tower-lightning wire based on the classified tower-lightning wire point cloud data using a material filtering algorithm. Specifically, this includes: calculating the ground point cloud data in the tower-lightning wire point cloud data using a material filtering algorithm; constructing a digital surface model based on the ground point cloud data; calculating the nearest neighbor point between the lightning wire and the tower to obtain the lightning wire attachment point; and calculating the lightning wire height parameters based on the projection of the lightning wire attachment point onto the digital surface model.

[0135] In one possible implementation, the processing module 32 is used to calculate the ground point cloud data in the tower-lightning wire point cloud data using a material filtering algorithm, specifically including: the material filtering algorithm is expressed by the following formula:

[0136] ;

[0137] in, Indicates that fabric particles are in Location at any given moment For the mass of the fabric particles, The external force acting on the fabric particles The internal forces acting on the fabric particles.

[0138] In one possible implementation, the processing module 32 is used to calculate the height parameters of each phase conductor of the tower based on the classified conductor tower point cloud data and through a digital terrain model. Specifically, this includes: calculating the nearest neighbor point between the conductor and the tower insulator to obtain the conductor suspension point; obtaining the relative position of the conductor suspension point based on the classified conductor tower point cloud data; and calculating the height parameters of each phase conductor of the tower based on the projection and relative position of the conductor suspension point on the digital surface model.

[0139] In one possible implementation, the processing module 32 is used to obtain the relative position of the conductor suspension point based on the classified conductor tower point cloud data, specifically including: obtaining the suspension point coordinates of each phase conductor on the insulator based on the conductor tower point cloud data; sorting the suspension point coordinates and defining the observer's perspective based on the sorting result; calculating the spatial relationship between the suspension point connection vector and the observation vector based on the observer's perspective; and obtaining the relative position of the conductor suspension point in the vertical and horizontal directions through the spatial relationship.

[0140] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0141] The present invention also provides an electronic device. (See reference...) Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device may include: at least one processor 401, at least one communication bus 402, a user interface 403, at least one network interface 404, and a memory 405.

[0142] The communication bus 402 is used to enable communication between these components.

[0143] The user interface 403 may include a display screen and a camera. Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0144] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0145] The processor 401 may include one or more processing cores. The processor 401 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 405, and by calling data stored in memory 405. Optionally, the processor 401 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 401 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 401.

[0146] The memory 405 may include random access memory (RAM) or read-only memory. Optionally, the memory 405 may include a non-transitory computer-readable storage medium. The memory 405 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 405 may also be at least one storage device located remotely from the aforementioned processor 401. (Refer to...) Figure 4 The memory 405, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for extracting lightning damage risk assessment calculation parameters.

[0147] exist Figure 4 In the illustrated electronic device, the user interface 403 is primarily used to provide an input interface for the user and acquire user-input data; while the processor 401 can be used to call the lightning damage risk assessment calculation parameter extraction application stored in the memory 405. When executed by one or more processors 401, the electronic device performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0148] The present invention also provides a computer-readable storage medium storing instructions. When executed by one or more processors, these instructions cause an electronic device to perform one or more of the methods described in the above embodiments.

[0149] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0150] In the various embodiments provided by this invention, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0153] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0154] The above description is merely an exemplary embodiment of the present invention and should not be construed as limiting the scope of the present invention. Any equivalent changes and modifications made in accordance with the teachings of the present invention shall still fall within the scope of the present invention.

[0155] This invention is intended to cover any variations, uses, or adaptations disclosed herein that follow the general principles disclosed herein and include common knowledge or customary techniques in the art not described in this invention.

Claims

1. A method for extracting calculation parameters for lightning damage risk assessment, characterized in that, The method includes: The laser point cloud data of each instantiated power component is acquired, and the laser point cloud data is classified according to the equipment type of each power component. The classified laser point cloud data includes insulator point cloud data, tower-lightning wire point cloud data, and conductor tower point cloud data. Based on the classified insulator point cloud data, the insulator string distance parameters are calculated using principal component analysis. Based on the classified tower-lightning conductor point cloud data, the tower-lightning conductor height parameters are calculated using a material filtering algorithm. The relative positions of the conductor suspension points are obtained based on the classified point cloud data of the conductor towers, and the height parameters of each phase conductor of the towers are calculated using a digital terrain model. Based on the insulator string distance parameter, the tower lightning protection wire height parameter, and the tower phase conductor height parameter, the lightning damage risk assessment calculation parameters are constructed.

2. The method according to claim 1, characterized in that, The process of acquiring laser point cloud data for each instantiated power component and classifying the laser point cloud data according to the device type of each power component specifically includes: Semantic segmentation processing is performed on the laser point cloud data to extract a subset of point clouds with semantic labels; Based on the aforementioned point cloud subset, an adjacency graph is constructed using k-nearest neighbor search; Based on the adjacency graph, the laser point cloud data of the same equipment type is divided into several connected graphs through connected component analysis, and one connected graph corresponds to a single power equipment in the equipment type; The connected graph is removed according to a preset method to obtain the classified laser point cloud data.

3. The method according to claim 1, characterized in that, The step of calculating the insulator string distance parameters using principal component analysis based on the classified insulator point cloud data specifically includes: The insulator point cloud data is standardized to obtain a data matrix corresponding to each insulator; Construct a covariance matrix based on the data matrix; The feature vector corresponding to the insulator point cloud data is obtained by solving the covariance matrix. Select a preset number of the feature vectors to construct the principal component data corresponding to the insulator point cloud data; The insulator point cloud data is projected onto the direction of the principal component data to obtain the projected point cloud, and the distance between the two endpoints of the projected point cloud is calculated. The distance between the two endpoints of the projected point cloud is used as the distance parameter of the insulator string.

4. The method according to claim 3, characterized in that, The construction of the covariance matrix based on the data matrix specifically includes: The covariance matrix is ​​constructed using the following formula: ； in, Let be the covariance matrix. For the data matrix, The number of points in the data matrix. It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

5. The method according to claim 1, characterized in that, The step of calculating the tower-lightning conductor height parameters using a material filtering algorithm based on the classified tower-lightning conductor point cloud data specifically includes: The ground point cloud data in the tower-lightning wire point cloud data is calculated using a fabric filtering algorithm; A digital land surface model is constructed based on the ground point cloud data; Calculate the nearest neighbor point between the lightning protection wire and the tower to obtain the lightning protection wire hanging point; The height parameter of the lightning protection wire is calculated based on the projection of the lightning protection wire attachment point onto the digital surface model.

6. The method according to claim 5, characterized in that, The fabric filtering algorithm is expressed by the following formula: ； in, Indicates that fabric particles are in Location at any given moment The mass of the fabric particles. The external force acting on the fabric particles. The internal force exerted on the fabric particles is denoted as .

7. The method according to claim 1, characterized in that, The step of calculating the conductor height parameters for each phase of the tower using a digital terrain model based on the classified point cloud data of the conductor towers specifically includes: Calculate the nearest neighbor points between the conductor and the tower insulator to obtain the conductor suspension points; The relative positions of the conductor suspension points are obtained based on the classified point cloud data of the conductor towers. The height parameters of each phase conductor of the tower are calculated based on the projection of the conductor suspension point on the digital surface model and the relative position.

8. The method according to claim 7, characterized in that, The step of obtaining the relative position of the conductor suspension point based on the classified conductor tower point cloud data specifically includes: Based on the point cloud data of the conductor tower, obtain the coordinates of the hanging points of each phase conductor on the insulator; The coordinates of the hanging points are sorted, and the observer's perspective is defined based on the sorting results; Calculate the spatial relationship between the hanging point connection vector and the observation vector based on the observer's perspective; The relative positions of the wire hanging points in the vertical and horizontal directions are obtained through the spatial relationship.

9. A device for extracting calculation parameters for lightning damage risk assessment, characterized in that, The device includes an acquisition module and a processing module, wherein, The acquisition module is used to acquire laser point cloud data of each instantiated power component, and classify the laser point cloud data according to the equipment type of each power component. The classified laser point cloud data includes insulator point cloud data, tower-lightning wire point cloud data and conductor tower point cloud data. The processing module is used to calculate the insulator string distance parameters using principal component analysis based on the classified insulator point cloud data; calculate the tower lightning conductor height parameters using a material filtering algorithm based on the classified tower-lightning conductor point cloud data; obtain the relative positions of conductor suspension points based on the classified conductor tower point cloud data, and calculate the conductor height parameters of each phase of the tower using a digital terrain model; and construct the lightning damage risk assessment calculation parameters based on the insulator string distance parameters, the tower lightning conductor height parameters, and the conductor height parameters of each phase of the tower.

10. The apparatus according to claim 9, characterized in that, The acquisition module acquires laser point cloud data for each instantiated power component and classifies the laser point cloud data according to the device type of each power component, specifically including: Semantic segmentation processing is performed on the laser point cloud data to extract a subset of point clouds with semantic labels; Based on the aforementioned point cloud subset, an adjacency graph is constructed using k-nearest neighbor search; Based on the adjacency graph, the laser point cloud data of the same equipment type is divided into several connected graphs through connected component analysis, and one connected graph corresponds to a single power equipment in the equipment type; The connected graph is removed according to a preset method to obtain the classified laser point cloud data.

11. The apparatus according to claim 9, characterized in that, The processing module calculates the insulator string distance parameters based on the classified insulator point cloud data using principal component analysis, specifically including: The insulator point cloud data is standardized to obtain a data matrix corresponding to each insulator; Construct a covariance matrix based on the data matrix; The feature vector corresponding to the insulator point cloud data is obtained by solving the covariance matrix. Select a preset number of the feature vectors to construct the principal component data corresponding to the insulator point cloud data; The insulator point cloud data is projected onto the direction of the principal component data to obtain the projected point cloud, and the distance between the two endpoints of the projected point cloud is calculated. The distance between the two endpoints of the projected point cloud is used as the distance parameter of the insulator string.

12. The apparatus according to claim 11, characterized in that, The construction of the covariance matrix based on the data matrix specifically includes: The covariance matrix is ​​constructed using the following formula: ； in, Let be the covariance matrix. For the data matrix, The number of points in the data matrix. It is a matrix composed of the mean values ​​of each feature dimension in the data matrix.

13. The apparatus according to claim 9, characterized in that, The processing module calculates the tower-lightning conductor height parameters based on the classified tower-lightning conductor point cloud data using a material filtering algorithm, specifically including: The ground point cloud data in the tower-lightning wire point cloud data is calculated using a fabric filtering algorithm; A digital land surface model is constructed based on the ground point cloud data; Calculate the nearest neighbor point between the lightning protection wire and the tower to obtain the lightning protection wire hanging point; The height parameter of the lightning protection wire is calculated based on the projection of the lightning protection wire attachment point onto the digital surface model.

14. The apparatus according to claim 13, characterized in that, The fabric filtering algorithm is expressed by the following formula: ； in, Indicates that fabric particles are in Location at any given moment The mass of the fabric particles. The external force acting on the fabric particles. The internal force exerted on the fabric particles is denoted as .

15. The apparatus according to claim 1, characterized in that, The processing module calculates the conductor height parameters for each phase of the tower using a digital terrain model based on the classified point cloud data of the conductor towers. Specifically, this includes: Calculate the nearest neighbor points between the conductor and the tower insulator to obtain the conductor suspension points; The relative positions of the conductor suspension points are obtained based on the classified point cloud data of the conductor towers. The height parameters of each phase conductor of the tower are calculated based on the projection of the conductor suspension point on the digital surface model and the relative position.

16. The apparatus according to claim 15, characterized in that, The step of obtaining the relative position of the conductor suspension point based on the classified conductor tower point cloud data specifically includes: Based on the point cloud data of the conductor tower, obtain the coordinates of the hanging points of each phase conductor on the insulator; The coordinates of the hanging points are sorted, and the observer's perspective is defined based on the sorting results; Calculate the spatial relationship between the hanging point connection vector and the observation vector based on the observer's perspective; The relative positions of the wire hanging points in the vertical and horizontal directions are obtained through the spatial relationship.

17. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 8.

18. A computer-readable storage medium storing a computer program, the computer program being executed by a processor according to any one of claims 1 to 8.

Images